Pupil Segmentation Using Texture and Intensity Combination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pupil segmentation methods struggle to accurately describe pupil shapes that deviate significantly from circular or elliptic shapes, which can lead to distortions in iris analysis due to irregular pupil dilation and contraction, affecting the robustness of biometric identification systems.
Innovation Solution
A method for pupil segmentation in digital images that involves deriving a texture image, forming a combined image with intensity and texture values, and approximating the pupil boundary with a convex curve, allowing for a more robust description of irregular pupil shapes and enabling effective normalization of iris images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional pupil segmentation methods (based on low intensity and low contrast) are used, then the method is simple and fast, but it cannot accurately describe pupil shapes that deviate significantly from circular or elliptic shapes
Solution Approach 1:
The patent segments the pupil boundary description into two parts: an approximate convex curve (ellipse or circle) for the overall shape, and convex polygons for the irregular extensions. This segmentation allows accurate representation of complex pupil shapes while maintaining a structured approach to the segmentation process.
Solution Approach 2:
The patent extends the traditional 2D boundary representation by adding a third dimension through convex polygons that represent irregular extensions. This dimensional enhancement allows the model to capture complex pupil shapes that cannot be described by simple convex curves alone.
2Reliability
If pupil segmentation accurately captures irregular shapes, then biometric identification reliability improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent performs preliminary segmentation to identify the approximate convex curve boundary first, then adds convex polygon extensions only where needed. This preliminary action approach reduces computational complexity by avoiding full complex boundary analysis when simple shapes suffice.
Solution Approach 2:
The patent applies different levels of description complexity to different regions of the pupil boundary. Simple convex curves are used for regular regions, while convex polygons are applied only to local irregular extensions. This local quality approach maintains reliability where needed while reducing overall processing time.
3Measurement precision
If a simple convex curve approximation is used, then processing is faster, but the pupil boundary accuracy deteriorates for irregular shapes
Solution Approach 1:
The patent employs a dynamic boundary description that adapts to the complexity of the pupil shape. The boundary model automatically adjusts between simple convex curve approximation and extended convex polygon representation based on the actual shape characteristics, optimizing the balance between accuracy and processing speed.
Solution Approach 2:
The patent applies the more complex convex polygon extension only partially, specifically to regions where irregular extensions are detected, rather than applying it uniformly to the entire boundary. This partial action approach maintains boundary accuracy where needed while preserving processing speed in regular regions.
Data Source
AI summary
A method of pupil segmentation in a digital image of a vertebrate eye, said image being an intensity image composed of pixels having each a specific intensity value, the method comprising the steps of:deriving a texture image from the intensity image, said texture image being composed of pixels having each a specific contrast value;forming a combined image by point-wise combining the intensity image with the texture image,identifying a set of pixels in the combined image which fulfill a combined low-intensity and low-contrast criterion; andapproximating a boundary of said set by a convex curve and taking said convex curve as a boundary of the pupil.